1998/03/31 by Esther M. Hu, Lennox L. Cowie, Richard G. McMahon · 1 citation
Physics and Astronomy · #astro-ph
paper · pdf · doi:10.1086/311506
published as Astrophys.J. 502 (1998) L99-L103 · 7 pages, 5 encapsulated figures; aastex, emulateapj, psfig and lscape style files. Separate gif files for 2 gray-scale images also available at http://www.ifa.hawaii.edu/faculty/hu/emitters.html . Added discussion of foreground contaminants. Updated discussion of comparison with external surveys (Sec. 5 and Fig. 5). Note: continuum break strength limits (Fig. 3 caption) are correct here -- published ApJL text has a sign error
We describe narrowband and spectroscopic searches for emission-line star forming galaxies in the redshift range 3 to 6 with the 10 m Keck II Telescope. These searches yield a substantial population of objects with only a single strong (equivalent width >> 100 Angstrom) emission line, lying in the 4000 - 10,000 Angstrom range. Spectra of the objects found in narrowband-selected samples at lambda ~5390 Angstroms and ~6741 Angstroms show that these very high equivalent width emission lines are generally redshifted Lyman alpha 1216 Angstrom at z~3.4 and 4.5. The density of these emitters above the 5 sigma detection limit of 1.5 e-17 ergs/cm2/s is roughly 15,000 per square degree per unit redshift interval at both z~3.4 and 4.5. A complementary deeper (1 sigma \~1.0 e-18 ergs/cm2/s) slit spectroscopic search covering a wide redshift range but a more limited spatial area (200 square arcminutes) shows such objects can be found over the redshift range 3 to 6, with the currently highest redshift detected being at z=5.64. The Lyman alpha flux distribution can be used to estimate a minimum star formation rate in the absence of reddening of roughly 0.01 solar masses/Mpc3/year (H0 = 65 km/s/Mpc and q0 = 0.5). Corrections for reddening are likely to be no larger than a factor of two, since observed equivalent widths are close to the maximum values obtainable from ionization by a massive star population. Within the still significant uncertainties, the star formation rate from the Lyman alpha-selected sample is comparable to that of the color-break-selected samples at z~3, but may represent an increasing fraction of the total rates at higher redshifts. This higher-z population can be readily studied with large ground-based telescopes.